In order to overcome the problem of virtual shadow at the edge of the target object in the process of infrared and visible image fusion, divide region universe algorithm is proposed. Firstly, the space model of the universe was established, universe in the core region was used two-way communication, while the universe in the non core region was used one-way communication only.Secondly, the connected universe was formed neighborhood relationship between the core region and the non core region. Thirdly, the detail layer, basic layer and salient layer were obtained from infrared and visible image respectively, and different fusion strategies were used for different layers.Finally, the fusion process of universe optimization was given. The experimental simulation shows that the fusion image result of divide region universe algorithm has more hot target and scene information, and the visual effect conforms to the characteristics of the human eye, average value of MI, SD, SF and AG is increased by 17.57%, 14.80%, 34.16% and 21.01% compared with other algorithms,it is better than other algorithms.